aaron-marketing-skills
Pass
Audited by Gen Agent Trust Hub on Jun 28, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill is designed to process external web content from URLs provided in commands such as /aaron-marketing:audit and /aaron-marketing:research. This creates an indirect prompt injection surface where instructions hidden in the target webpage's HTML or metadata could attempt to override the agent's behavior or exfiltrate data.
- Ingestion points: External URLs are ingested by several skills, including content-quality-auditor, on-page-seo-auditor, and domain-authority-auditor in SKILL.md.
- Boundary markers: No explicit markers (like XML tags or specific delimiters) or instructions to ignore embedded commands are documented for the processing of external content.
- Capability inventory: The skill has the ability to execute local shell scripts (via commands/ scripts) and likely performs network requests to fetch content from the provided URLs.
- Sanitization: There is no documentation regarding the sanitization, filtering, or validation of the content retrieved from external sources before it is processed by the AI agent.- [EXTERNAL_DOWNLOADS]: The documentation includes instructions to download the skill and its components from the author's GitHub repository (aaron-he-zhu/aaron-marketing-skills) using standard tools like git, npx, and the Claude Code plugin manager. These installation patterns are consistent with the intended delivery mechanism for this toolset.- [COMMAND_EXECUTION]: The skill uses a custom framework (run_skill) and Bash wrappers to execute local scripts and Python-based logic for various marketing tasks. This involves running executable files within the project structure (e.g., ./skills/optimize/on-page-seo-auditor/run.sh), which is the standard operational model for this collection of skills.
Audit Metadata